| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
ORIGINAL RESEARCH COMMUNICATION |
1 From the Department of Primary Care and Population Sciences, Royal Free and University College Medical School, London, United Kingdom (SGW, AGS, and LL), and the Division of Community Health Sciences, St George's, University of London, London, United Kingdom (PHW)
2 Supported by the British Heart Foundation. 3 Reprints not available. Address correspondence to SG Wannamethee, Department of Primary Care and Population Sciences, Royal Free and University College Medical School, Rowland Hill Street, London NW3 2PF, United Kingdom. E-mail: goya{at}pcps.ucl.ac.uk.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Objective: We examined the relation between other anthropometric indexes of body composition (both muscle mass and body fat) and all-cause mortality in men aged 60–79 y.
Design: The study was a prospective study of 4107 men aged 60–79 y with no diagnosis of heart failure and who were followed for a mean period of 6 y, during which time there were 713 deaths.
Results: Underweight men (BMI < 18.5) had exceptionally high mortality rates. After the exclusion of these men, increased adiposity [BMI, waist circumference (WC), and waist-to-hip ratio] showed little relation with mortality after adjustment for lifestyle characteristics. Muscle mass [indicated by midarm muscle circumference (MAMC)] was significantly and inversely associated with mortality. After adjustment for MAMC, obesity markers, particularly high WC (>102 cm) and waist-to-hip ratio (top quartile), were associated with increased mortality. A composite measure of MAMC and WC most effectively predicted mortality. Men with low WC (
102 cm) and above-median muscle mass showed the lowest mortality risk. Men with WC > 102 cm and above-median muscle mass showed significantly increased mortality [age-adjusted relative risk: 1.36; 95% CI: 1.07, 1.74), and this increased to 1.55 (95% CI: 1.01, 2.39) in those with WC > 102 and low MAMC.
Conclusion: The findings suggest that the combined use of both WC and MAMC provides simple measures of body composition to assess mortality risk in older men.
Key Words: Body mass index • waist circumference • muscle mass • adiposity • mortality
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
25) and obesity (30) irrespective of age (1, 2). The importance of overweight and obesity as predictors of mortality in elderly persons (>65 y) is controversial (3-5). In Western societies, BMI tends to increase with age up to 
65 y and then decrease (6). Whereas a few studies have shown U-shaped or positive associations between BMI and mortality in the elderly (7-9), most of the studies in the elderly have shown no significant increase in mortality in those who are overweight or obese (3-5, 10-19). Some studies have even reported an inverse association between BMI and mortality, in which mortality risk is lower for subjects with an elevated BMI than for leaner subjects (3-5, 13-19). Aging is associated with significant changes in body composition, with a substantial reduction in fat-free mass (FFM) and muscle mass and an increase in visceral fat, even if body weight remains unchanged (20-22). Thus, BMI depends not only on adiposity but on the loss of muscle mass, which has opposing effects on mortality (23-25) so that BMI may not be a good indicator of obesity in the elderly. It is suggested that measures of adiposity such as waist circumference (WC) or the waist-to-hip ratio (WHR), which better reflect visceral fat (26-27), may be better indicators of mortality in the elderly than is BMI (12, 13). Several studies have shown high WC, high WHR, or an increased percentage body fat to be associated with increased mortality in older subjects (12, 13, 17, 28). We have therefore assessed whether other indexes of body composition, which include markers of adiposity [WC, WHR, and fat mass (FM)] and markers of muscle mass [FFM and midarm muscle circumference (MAMC)], better explain mortality risk in the elderly than does BMI. We also assessed the combined use of markers of adiposity and muscle mass as predictors of mortality. |
SUBJECTS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Ethical approval was provided by all relevant local research ethics committees. All of the men provided written informed consent to the investigations, which were carried out in accordance with the Declaration of Helsinki.
Anthropometric measurements
Weight and height were measured at the initial examination (1978–1980). Measurements at reexamination (1998–2000) included height, weight, waist and hip circumferences, FFM, triceps skinfold thickness, and midupper arm circumference (MUAC). Subjects were measured standing and wearing light clothing without shoes. Height was measured with the use of a stadiometer (Holtain, Crosswell, United Kingdom) to the last complete 0.1 cm, and weight was measured with the use of a digital electronic scale (Soehnle-Waagen, Murrhardt, Germany) to the last complete 0.1 kg. BMI was calculated for each man. Waist and hip circumferences were measured in duplicate with the use of an insertion tape (CMS Ltd, London, United Kingdom); hip circumference was measured at the point of maximum circumference over the buttocks. The waist measurement was taken from the midpoint between the iliac crest and the lower ribs measured at the sides. The midupper arm was defined, and MUAC was measured once with the use of the arm pendant. Triceps skinfold thickness was measured at the midupper arm in duplicate, and the average of the 2 readings was taken. WHR was calculated as WC divided by hip circumference (cm). FM and FFM were calculated by using bioelectrical impedance analysis with the use of a Bodystat 500 apparatus (Bodystat Ltd, Douglas, United Kingdom). FM was calculated by using the equation by Deurenberg et al (31), which is designed for persons aged >60 y. FFM was calculated as 6710 x height (in m)2/resistance (in 
) + 7. FM was calculated as body weight –FFM. To allow the comparison of subjects with different heights, FM and FFM measures were normalized for height by dividing them by (height)2 to obtain the FM index and the FFM index (32). MAMC was calculated as MUAC – 0.3142 x (triceps skinfold thickness) (33). WC, WHR, and FM were used as indicators of adiposity; FFM and MAMC were considered indicators of muscle mass. Height loss was defined as loss in height over 20 y from the initial examination to the rescreening. Those who had a loss 3 cm were considered to have significant height loss. Weight loss was defined as a loss of 4% in weight (34) during that period.
The NIH and the World Health Organization classifications of overweight (BMI 25–29.9) and obesity (BMI 30) were used (1-2). Within the overweight category, the men were divided into 2 groups to assess whether any differences existed between men in the lower and the upper ranges of the overweight category. The men were classified into 5 groups on the basis of their current BMI: <18.5,18.5–24.9 (normal weight), 25–27.4 and 27.5–29.9 (overweight), and >30 (obese). For WC, the men were divided into 4 groups on the basis of the levels associated with increased cardiovascular disease proposed by Lean et al (35): <94, 94–102, 103–106, and >106 cm). High WC was defined as >102 cm, which is the threshold cutoff for WC used by the NIH (2). For WHR, FM, FFM, and MAMC, the men were divided by quartiles of the distribution.
Cardiovascular disease risk factors
The measurement and classification methods for smoking status, physical activity, social class, and blood pressure in this cohort were described previously (29, 36, 37). From the combined information at the initial screening (Q1; 1978–1980), follow-up questionnaires in 1996 (Q96), and at rescreening (Q20), the men were classified into 5 smoking groups: 1) those who had never smoked, 2) exsmokers since screening, 3) smokers at baseline who gave up between screening and Q96, 4) smokers at screening and Q96 who gave up after 1996, and 5) current cigarette smokers at Q20. On the basis of the frequency and type of activity, a physical activity score was derived for each man, and the men were grouped into 6 broad categories: inactive, occasional, light, moderate, moderately vigorous, and vigorous. Men who reported being inactive or occasionally active were referred to as "inactive." Forced expiratory volume in 1 s (FEV1) was used to measure lung function and was height-standardized to the average height, 1.73 m, in the study. Thus, height-standardized FEV1 = FEV1 x (1.73/height)2. The glomerular filtration rate (eGFR) estimated from serum creatinine by using the Modification of Diet in Renal Disease equation developed by Levy et al (38, 39) was used as a measure of renal function. C-reactive protein (CRP) was assayed by using ultrasensitive nephelometry (Dade Behring, Milton Keynes, United Kingdom).
Follow-up
All men were followed from initial examination (1978–1980) to December 2005 for all-cause mortality (40), and follow-up has been achieved for 99% of the cohort. Information on death was collected through the established "tagging" procedures provided by the National Health Service registers (40). In the present analyses, all-cause mortality was based on follow-up from the rescreening in 1998–2000 at the mean age of 60–79 y, which is a mean follow-up period of 6 y (range: 5–7 y).
Statistical methods
Statistical analyses were performed by using SAS software (version 9.2; SAS Institute Inc, Cary NC). Cox's proportional hazards model was used to assess the adjusted relative risk (RR) for the adiposity variables. In the adjustment, smoking, social class, alcohol intake, physical activity, weight loss, height loss, and the presence of diseases were fitted as categorical variables. Age, FEV1, albumin, and CRP were fitted as continuous variables.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Body mass index and mortality
The relation between BMI at rescreening and mortality over the 6-y mean follow-up period is shown in Table 1
. Underweight men had by far the highest mortality. The lowest mortality was seen in the overweight men (BMI 25–29.9) after adjustment for age and lifestyle characteristics (model 2). Obese men showed risk similar to that of normal-weight men. Adjustment for indicators of ill health predictive of mortality and associated with leanness and smoking (low albumin, lung function, stroke, and cancer) made minor differences to the risk seen in obese men but attenuated the lower risk seen in overweight men, and the difference was no longer significant (model 3). The exclusion of current smokers and recent exsmokers made minimal difference (model 4).
|
), excluding the 24 underweight men. In the age-adjusted analyses, significant positive associations were seen between WC, WHR, and mortality, which were attenuated after adjustment for lifestyle characteristics (model 2). No associations were seen with FM index and FFM index. MAMC showed a significant inverse association with mortality even after these adjustments.
|
). Forty-seven percent of normal-weight men had low MAMC (lowest quartile), whereas 13.3% of overweight men and 5% of obese men had low MAMC. The association between WC, WHR, and mortality became significantly positive after adjustment for MAMC (Table 2
), as did the association between BMI and mortality. The adjusted RRs (and 95% CIs) for the 4 BMI groups (excluding underweight men) were 1.00, 0.99 (0.72, 1.08), 0.96 (0.76, 1.22), and 1.26 (0.97, 1.63) (P for trend = 0.03).
|
. The percentage of men who lost substantial height (3 cm) increased progressively and significantly with decreasing MAMC, from 9.6% in the top quartile to 20.8% in the lowest quartile. Over 30.8% of men in the lowest MAMC quartile had weight loss over the past 20 y, whereas only 6.6% in the top quartile had weight loss. MAMC correlated positively with FEV1 and albumin and was inversely associated with height loss, all of which are factors shown to predict mortality. However, an inverse association was seen with renal function. High rather than low MAMC was associated with lower eGFR; low eGFR is associated with increased cardiovascular disease mortality (39). No significant association was seen with CRP. The association between MAMC and mortality persisted even after adjustment for indicators of ill health, which include FEV1, albumin, weight loss, height loss, self-reported poor health, preexisting cancer, diabetes, and cardiovascular disease. The adjusted RRs (and 95% CIs) for the 4 quartiles were 1.00, 0.83 (0.67, 1.02), 0.73 (0.58, 0.92), and 0.73 (0.58, 0.92), respectively (P for trend = 0.002). Further adjustment for CRP made little difference to the findings. The inverse association was similar in both physically inactive and active men (P for trend = 0.008 and 0.01, respectively).
Adiposity, midarm muscle circumference, and mortality
The relation between adiposity measures (BMI and WC) and mortality stratified by the 4 MAMC groups is shown in Figure 1. Men with low MAMC (lowest quartile) showed the highest risk irrespective of BMI. BMI was positively associated with mortality in men with higher MAMC (top 2 quartiles); an inverse association was seen in those with below-median MAMC (P < 0.0001 test for interaction). Within all MAMC groups, WC was positively associated with mortality, although the associations were stronger in those with higher MAMC (P = 0.005 test for interaction; Figure 1). The adjusted RRs for BMI and WC stratified by MAMC with the highest 2 MAMC quartiles combined is shown in Table 4. After adjustment, WC was positively associated with mortality in those with medium or high MAMC; no association was seen in the low-MAMC group. BMI was significantly positively associated with mortality in men with higher MAMC (above median).
|
|
Combined use of muscle arm circumference and high waist circumference
The combined effect of high WC and MAMC on mortality risk with adjustment for age, social class, and other lifestyle characteristics is shown in Table 5; men with higher MAMC and WC < 102 cm were used as the reference group. Those with high MAMC and WC < 102 cm had the lowest mortality rates, and those with below-median MAMC and WC had the highest rates. After adjustment for age and other characteristics, men with high MAMC and high WC showed a 36% increase in mortality risk (adjusted RR: 1.36; 95% CI: 1.07, 1.74), and the risk of mortality increased to nearly 60% in those with low MAMC and increased WC (adjusted RR: 1.55; 95% CI: 1.01, 2.39), although the highest risk was seen in those with medium MAMC (second quartile) and high WC after these adjustments.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
30) as defined by the NIH guidelines was associated with increased mortality as was normal-weight BMI (18.5–24.9). These findings are consistent with other population studies carried out in older persons (3-5, 10-19). More important, our results suggest that both increased adiposity as measured by high WC and low muscle mass as measured by MAMC independently predispose to mortality in older men and that the use of both these measures is necessary in assessing body composition and mortality in the elderly.
Body mass index and muscle mass
BMI correlated positively and strongly with MAMC, with almost one-half of normal-weight men having MAMC in the lowest quartile, which reflected low muscle mass. The relation between BMI and mortality appeared to be dependent on muscle mass, with obesity being positively associated with increased mortality in those with above-median MAMC. In particular, normal-weight men with higher MAMC did not show increased mortality. Thus, the increased risk of mortality associated with normal body weight (BMI 18.5–24.9) in the elderly may be due to the loss of muscle mass rather than to low body weight per se.
Abdominal fat distribution
Whereas some studies have shown WC or WHR to be better (positive) predictors of mortality than is BMI (12, 13, 15, 17), others have reported no association (10, 41). Most previous studies have not taken muscle mass into account in the examination of the measures of central adiposity and mortality. In the present study, weak (positive) associations were seen between WC and WHR with mortality, but these became significantly positive after adjustment for MAMC, with stronger positive associations seen between WC and mortality than between WHR and mortality. The positive association between WC and mortality was not apparent in those with low MAMC. The reason for this finding is not clear, but it is possible that those with low MAMC (possibly reflecting low muscle strength, ill health, or loss in muscle mass) are at such increased risk of mortality that adiposity has little influence on mortality in these men. The association between WC and mortality persisted even after adjustment for BMI; in contrast, no association was seen between BMI and mortality after adjustment for WC, which suggests that WC is a better indicator of mortality than is BMI in older men, which is in keeping with other reports (12, 13, 15, 17).
Muscle mass and mortality
The finding of a strong inverse association between MAMC and mortality is consistent with other prospective studies linking mortality to MAMC (33), FFM (23), or the sum of midarm and midthigh circumference (24). In the present study, low MAMC was strongly associated with weight loss, which possibly reflects loss in muscle mass, and appeared to be a stronger predictor of mortality than was increased adiposity. Men in the lowest quartile of the MAMC distribution showed increased mortality irrespective of adiposity level, which emphasizes the importance of the preservation of muscle mass in the elderly. The increased mortality associated with low MAMC was not explained by indicators of ill health or inflammation (CRP). An inverse association was seen with renal function (eGFR); thus, adjustment for renal function, if anything, would strengthen the increased risk. The loss of muscle mass results in reduction in muscle strength with increasing age, which leads to impaired mobility (42, 43). Muscle strength is closely related to muscle mass and has been found to be a strong predictor of mortality (44-46). We cannot say whether a reduction in muscle mass is a reflection of ill health or whether the associated loss of muscle strength and impaired mobility has untoward consequences.
Fat mass and fat-free mass
Although several studies have shown FM and FFM as measured by the bioelectric impedance method to have opposing effects on mortality (23, 25), most of these studies were carried out in a younger population (aged <65 y). The weak association seen between the bioelectric impedance measure of FM and FFM with mortality in the present study may have been due to imprecision. It has been suggested that, although bioelectric impedance allows a better measurement of body composition than does weight, it is limited by assumptions of water distribution and is dependent on population-specific equations for analysis (31). The impedance measurements may overestimate FFM in elderly individuals and thereby underestimate body fat (47). We also calculated FFM and FM by using the equation developed by Sun et al (48), which was validated in a US cohort of men and women aged 12–94 y. This equation yielded similar findings.
Composite use of waist circumference and midarm muscle circumference
It is well recognized that BMI is an inaccurate measure of body fatness in the elderly because of the decrease in skeletal muscle mass and the increase in abdominal fat with aging (21-22). Our findings suggest that the combined use of WC and MAMC provided the best indexes of body composition to assess mortality risk in the elderly. Such assessment, although it would include observer-dependent measurements of WC, MUAC, and triceps skinfold, would be feasible with appropriate training. Measurement of WC is already being increasingly recommended in clinical practice (49). Increased adiposity (WC > 102 cm) or low muscle mass as defined by the lowest quartile of MAMC were both associated with significantly increased mortality in the elderly; mortality was lowest in those with above-median MAMC and low WC and was highest in those with below-median MAMC and increased WC. Thus, central fat and relative loss of muscle mass may become relatively more important than BMI in determining health risk associated with obesity in older age.
Limitations of the study
Our study was carried out in a population-based study of men. Although it is possible that these findings would apply in women, such generalization must be cautious. The presence of doctor-diagnosed diseases was based on self-reporting, and we did not have information on all comorbidity that may have related to low muscle mass. However, we showed the increased risk of mortality associated with MAMC to be seen still after adjustment for several objective markers of ill health, such as weight loss, lung function, albumin, and inflammation. Although we had no direct measures of visceral fat, WC has been shown to correlate strongly with more precise measures of visceral fat as assessed by computed tomography scan in the elderly (50). We did not have direct measures of muscle strength and skeletal muscle mass; FFM, as assessed by bioelectrical impedance methods, is not the most precise method for measuring skeletal muscle mass.
In conclusion, increased central adiposity and decreased muscle mass are associated with increased mortality. Body-composition assessment is more informative than are simple body weight measurements in predicting mortality in the elderly. Although BMI should be used in the first instance to identify underweight men, our findings suggest that the combined use of WC as a marker of central adiposity and of MAMC as a marker of muscle mass provides simple measures of body composition in assessing mortality risk in elderly men.
|
ACKNOWLEDGMENTS |
|---|
The authors' responsibilities were as follows—SGW: the study concept and data analysis and drafting the manuscript; AGS: the design of the original study; PHW: the 20-y rescreening of the study population; AGS and PHW: drafting the manuscript; and LL: the follow-up of the men. None of the authors had a conflict of interest.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  O. Bouillanne, C. Dupont-Belmont, P. Hay, B. Hamon-Vilcot, L. Cynober, and C. Aussel Fat mass protects hospitalized elderly persons against morbidity and mortality Am. J. Clinical Nutrition, September 1, 2009; 90(3): 505 - 510. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. W. Auyeung, J. S. W. Lee, J. Leung, T. Kwok, P. C. Leung, and J. Woo Survival in Older Men May Benefit From Being Slightly Overweight and Centrally Obese--A 5-Year Follow-up Study in 4,000 Older Adults Using DXA J Gerontol A Biol Sci Med Sci, July 23, 2009; (2009) glp099v1. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. E. Alley, L. Ferrucci, M. Barbagallo, S. A. Studenski, and T. B. Harris A Research Agenda: The Changing Relationship Between Body Weight and Health in Aging J. Gerontol. A Biol. Sci. Med. Sci., November 1, 2008; 63(11): 1257 - 1259. [Full Text] [PDF]
|
|
|
|
 |
|
 M. L. Simoons and L. Bonneux Obesity, Cardiology, and Beyond J. Am. Coll. Cardiol., September 16, 2008; 52(12): 986 - 987. [Full Text] [PDF]
|
|
|
|
 |
|
 R. Ness-Abramof and C. M. Apovian Waist Circumference Measurement in Clinical Practice Nutr Clin Pract, August 1, 2008; 23(4): 397 - 404. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. B. Horwich and G. C. Fonarow Measures of Obesity and Outcomes After Myocardial Infarction Circulation, July 29, 2008; 118(5): 469 - 471. [Full Text] [PDF]
|
|
|
|
|
|
E. Atlantis, S. A Martin, M. T Haren, A. W Taylor, G. A Wittert, and for the Florey Adelaide Male Aging Study Lifestyle factors associated with age-related differences in body composition: the Florey Adelaide Male Aging Study Am. J. Clinical Nutrition, July 1, 2008; 88(1): 95 - 104. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Peeters BMI and cardiorespiratory fitness predicted mortality in older adults Evid. Based Med., June 1, 2008; 13(3): 90 - 91. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
